无知几何：LLM如何调节贝叶斯先验
原标题：The Geometry of Ignorance: LLMs Know When to Temper Bayesian Priors
AI 摘要
该研究揭示了大型语言模型（LLM）在不确定时如何依赖训练语料的词频先验：其输出嵌入矩阵中存在一个单一方向，编码了语料库的unigram分布，称为“无知方向”。该方向在Llama、Qwen、Gemma和Pythia等模型家族中普遍存在，且通过线性最小二乘拟合即可恢复。研究发现，将最终预测状态投影到该方向可分解为tempered Bayesian更新，其中先验加载因子随上下文信息量变化，且该方向具有因果活性，干预可改变预测。
正文节选
The Geometry of Ignorance: LLMs Know When to Temper Bayesian Priors Abstract What does a language model predict when it has few clues? The answer lurks in its unembedding geometry: a single direction of the unembedding matrix encodes the unigram distribution of the training corpus, which serves as the Bayesian prior the model falls back on when uncertain. This structure — which we term the direction of ignorance — appears in all four model families examined (Llama, Qwen, Gemma, and Pythia), rang